Re: IEEE488 electrical specs

From: Martijn van Buul (
Date: 2001-04-20 12:52:29 wrote:
> The specs of TI's SN75160B and SN75161B say a 1.7 K resistor plus a diode 
> for every output pin. Less then I thought :( 
> (5 - 0.6) / 1700 = 2.6 mA. So 6 devices shouldn't be a problem

Hmm. I'm having some problems, though; I intended to use P0 for the
8 datalines, but even *reading* them completely crashes the bus, and
often even resets the 8051. Very odd; P0 is a honest-to-heart open-collector
thing (doesn't even have internal pull-ups).

Hmm. Unless it's configured as push-pull for some obscure reason.

Anyway; I now use P2 for the datalines[1], and that seems to be working
like a charm. Stretching the timing of the bus works like a charm :)

By the way, does anybody know when ATN is valid (or to be more precise:
when it is *no longer* valid)?

> > Programming a 8751 isn't really difficult. I have the specifications 
> > floating somewhere on my desk at Philips. Interested in a copy? 
> Unfortunally it seems I don't have a 8751 although I still believe I have 
> at least one (but where ???). 

8751's are "common as muck". Hfl 5,-- at Display electronica 
(and you actually get a 8752, which has more RAM and timers..)

> On the other hand I found some 8741's, 8742's, 8748's and 8749's. I have 
> the Philips databook IC14 which contains info about the 8048 and 8049. But 
> NOT how to program them. I also have some 8048's but have no idea how to 
> find out if they are programmed or not.

Hmm. I might be able to find this info as well; Philips's intranet
is a *nightmare* though.

> AFAIK I don't have anything about the 8741 and 8742. If you have any info 
> on these, I would be glad if you could help me with the specs of these, 
> please. 

At second thought, I might have that particular info at home. I collected
a few databooks a few weeks ago, which I haven't really checked yet..

> I know the ML for these and the 8x51 differ but that shouldn't be a problem 
> if you would be so kind to donate the sources.

I'll try to keep that in mind when implementing the IEC-part of the device;
8041-derivates have much simpler timers and all :(

I don't know the 8041 quite well, but I seem to recall that it only had
one (external) interrupt line. The 8051 has two of them, and I intended
to use both of them (so that I can simply see if I got an 'IEC-ATN' or 
an 'IEEE' one :)

> Looking for the 8751 I bumped in something else from the time I worked at 
> NatLab: two 8051 PODs, a V85C552 and a V85C582. Interested?

You lost me somehwere; what's a POD? It sounds like In-cicuit emulators

> I also found a loose 80552 which reminded me I have an Elektuur-card 
> equiped with this controller. IMHO this one could be used as well. I only 
> hope I have enough I/O-pins left for the interface self is this card is 
> equiped with an EPROM and SRAM.

Hmm. I don't have my IC20-handbook at hand, but didn't the 552 have *6*
I/O ports?  Shouldn't be much of a problem, really.

I'm not certain external RAM is really required; I haven't used a single
byte of Data-space up to now (save stack)

> > It would be a shame if the connection between the PET and the 4040 would 
> > be slowed down because of the 1541..
> You would loose some time in the initial phase where all devices have to 
> check if the given devicenumber is theirs. It shouldn't slow down the 
> actual data transfers.

You're absolutely right. 

> For IEEE you need 8 datalines and 6 controllines (REN and SQR not needed), 
> for IEC only three: DATA, CLK and ATN. RESET is covered by IFC. That makes 
> 17. Two for simple RS232 makes 19. Five left which enables you 8K of EPROM. 
> Which is a lot, I thought. 

That's right :)

> > Maybe not; maybe I can abuse the UART for that..
> Very smart indeed, petje af :) 

I'm not quite sure if it will be worth the effort; the 8051's UART can't
be programmed very accurately at "high" bitrates, and I came up with a
good use for the UART - My mac is still missing some way of accessing 
CBM-drives; it might be a nice idea to implement something like 
'IEEE-over-RS232' :)

> But if RS232 is possible, why not. I programmed more then just a little bit 
> 8051 ten years ago. (the above Elektuur 80552 I used for an electronic 
> ignition for a VW Beetle with on-the-fly tuning while driving it on a 
> testbank) If I remembered well, the RS232 communication, once started, 
> didn't need any further attention.

Well, it only has a 1-byte "FIFO".. 

Ah well - I'll get back to it next monday; I'm going away for a weekend. 
Funny how these things always collide ;)


[1] Which is a bit of a nuisance; due to a bug^Wfeature of the 89c81Rx+
    processors, pins P2.6 and P2.7 *must* be floating when doing 
    in-circuit programming, which means that I have to disconnect the
    IEEE bus now :(. P3 got a 'stuck' bit somewhere (and it got the RS232
    interface on it), so that isn't useful anyway :(

    Martijn van Buul - -
	 Geek code: G--  - Visit OuterSpace: 3333
   Kees J. Bot: The sum of CPU power and user brain power is a constant.
This message was sent through the cbm-hackers mailing list.
To unsubscribe: echo unsubscribe | mail

Archive generated by hypermail 2.1.1.